Power customer portrait management method and system based on big data

Abstract

The invention provides a power customer portrait management method and system based on big data, and the method comprises the steps: collecting power operation data and environment monitoring data in real time through a power Internet of Things terminal, and accessing a social attribute database and a user behavior log server, establishing a comprehensive database to carry out first image simulation on each user to obtain a simulated image corresponding to each user; obtaining the appeal content of the current user, obtaining a simulated portrait corresponding to the current user, and recording the simulated portrait as a first portrait; judging whether the appeal content falls into the first portrait or not; if not, performing second simulation aiming at the current user to solve the appeal content; performing real-time reconstruction and quarterly reconstruction in combination with the second portrait, replacing the first portrait with a third portrait, and performing follow-up of subsequent appeals on the current user through the third portrait; and if yes, replying and solving the appeal content according to measures recorded by the first portrait. The intelligent degree of power customer portrait management is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power customer profiling, and more particularly, to a method and system for managing power customer profiling based on big data. Background Art

[0002] With the continuous deepening of the intelligent transformation of the power industry, the traditional power customer management mode has been difficult to meet the demand for refined services. At present, with the increasing demand for electricity, power customers also pay more and more attention to the speed of problem feedback being solved. At the same time, when maintaining existing power customers, more detailed and comprehensive services are also required. However, in the existing technology, customer profiles are mainly constructed by collecting basic power data such as user electricity consumption and payment records, resulting in problems such as single data dimension and one-sided feature extraction. Summary of the Invention

[0003] Therefore, the embodiments of the present invention provide a method and system for managing power customer profiling based on big data, which improves the intelligent level of power customer profiling management.

[0004] To solve the above problems, the present invention provides a method for managing power customer profiling based on big data, including: real-time collecting power operation data and environmental monitoring data through a power Internet of Things terminal, and at the same time accessing a social attribute database and a user behavior log server, and then establishing a comprehensive database; obtaining power data, environmental data, social data, and interaction data of each user according to the comprehensive database; performing a first simulation on each user according to the power data, environmental data, social data, and interaction data to obtain a corresponding simulation profile for each user; obtaining the content of the current user's appeal, obtaining the corresponding simulation profile of the current user and recording it as the first profile; determining whether the content of the appeal falls into the first profile; if not, performing a second simulation on the current user to obtain a second profile, and solving the content of the appeal according to the second profile; combining the second profile, performing real-time reconstruction and quarterly reconstruction on the first profile to obtain a third profile, and using the third profile to replace the first profile, and following up the subsequent appeals of the current user through the third profile; if so, replying and solving the content of the appeal according to the measures recorded in the first profile.

[0005] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: by collecting the power data, environmental data, social data, and interaction data of each user, the accuracy and comprehensiveness of the subsequent user portraits can be improved through multiple dimensions. The multi-dimensional approach can also facilitate dealing with different customer situations in the future, thereby better serving the customers. At the same time, when the current proposed portrait does not meet the customer's requirements, a secondary portrait is set up, and the customer's portrait is further improved through the secondary portrait method, so that better services can be provided to the customer when encountering the customer next time. In addition, real-time reconstruction and quarterly reconstruction are carried out to dynamically update the portrait, ensuring the real-time effectiveness and accuracy of the portrait, and thus better serving the customer in the future and managing the customer portrait more comprehensively. It can ensure better solving the problems requested by the customer, improving the accuracy and practicality of the portrait, and making the management method more intelligent.

[0006] In an example of the present invention, a first proposed image of each user is made according to the power data, environmental data, social data, and interaction data, and the proposed portrait corresponding to each user is obtained, including: the power data includes fault records, the environmental data includes meteorological information, the social data includes electricity bill information, and the interaction data includes customer service records; the proposed portrait of each user is made by combining the fault records, meteorological information, electricity bill information, and customer service records corresponding to each user.

[0007] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: through the fault records corresponding to each user, it is possible to clearly know the content of the user's previous fault reports, and thus obtain the problems that the user often needs to face and solve, which is convenient for subsequent portrait making. Through the meteorological information, it is possible to clearly know whether the user's location encounters bad weather that is likely to cause power failures, so as to better help judge the problems encountered by the user when reporting faults, ensure quick identification of problems and serve the user. At the same time, setting the electricity bill information can quickly judge the user's electricity bill situation, thus avoiding power problems caused by electricity bill problems. Furthermore, the customer service record is set as a factor, and the customer service record can reflect most of the user's demands, so that the portrait can be more accurate after combination.

[0008] In an example of the present invention, making the proposed portrait of each user by combining the fault records, meteorological information, electricity bill information, and customer service records corresponding to each user further includes: classifying the faults in the fault records and sorting them by the number of records to obtain a first sequence; combining the meteorological information and the first sequence to extract the first proposed image condition; combining the electricity bill information and the first sequence to extract the second proposed image condition; combining the customer service record and the first sequence to extract the third proposed image condition; making the proposed portrait of each user according to the first, second, and third proposed image conditions.

[0009] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: By setting the classification of fault records and then performing priority sorting, the faults of users can be quickly integrated, the problems frequently encountered by users can be identified, and then label identification and establishment can be quickly carried out, so as to better serve users. At the same time, different virtual image conditions are extracted by combining meteorological information, electricity bill information and customer service records respectively, so that the finally obtained virtual portrait can be more accurate and comprehensive, and the integrity of the portrait can be guaranteed.

[0010] In an example of the present invention, when extracting the first virtual image condition by combining meteorological information and the first sequence, it further includes: judging whether there is a power failure fault in the first sequence; if it exists and ranks among the top two in the first sequence, then combining meteorological information to judge whether the power failure is caused by bad weather; if it is judged to be the case, then add that the power failure is affected by bad weather, and a preventive measure label for the corresponding user needs to be informed in advance when encountering bad weather; if it is judged not to be the case, then add a measure label for line inspection in the area where the corresponding user is located.

[0011] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: By extracting the power failure fault and combining meteorological information to judge the specific cause of the power failure fault, preventive measure labels can be quickly added to users, and then users can be informed in advance when there will be bad weather in the future area where the users are located, improving the risk resistance ability of users and better serving users. At the same time, when it is not affected by bad weather, a label for line inspection in the area is added for users, so as to further avoid the power failure risk of users and ensure the normal power consumption of users.

[0012] In an example of the present invention, when extracting the second virtual image condition by combining electricity bill information and the first sequence, it further includes: judging whether there is an individual business power outage fault in the first sequence; if it exists and ranks among the top two in the first sequence, then combining electricity bill information to judge whether the power outage is caused by insufficient electricity bill balance; if so, then add a measure label for reminding the corresponding user to pay the electricity bill; if not, then add a measure label for line inspection in the area where the corresponding user is located.

[0013] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: By setting the judgment of the individual business power outage fault in combination with the electricity bill balance, it is possible to prevent users from having a power outage fault due to failure to submit the electricity bill in time, and then remind users early, so that users can pay the electricity bill in time, thus ensuring the normal power consumption of users.

[0014] In an example of the present invention, when extracting the third virtual image condition by combining customer service records and the first sequence, it further includes: judging whether the user needs recorded in the customer service records exist in the first sequence; if it exists and ranks among the top two in the first sequence, then generate a label for the solution corresponding to the demand record, and provide follow-up service to the user corresponding to the demand record.

[0015] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: By extracting the content in the customer service record and comparing it with the first sequence, it is possible to add records of the demands that users ask the customer service more frequently, and generate corresponding tags for the solutions provided by the customer service at that time. Thus, it is convenient to directly give an intelligent reply when encountering the same problem later, and subsequent follow-up services can be carried out to ensure the electricity consumption needs of users.

[0016] In an example of the present invention, if not, a second avatar is created for the current user to obtain a second portrait. Solving the demand content according to the second portrait further includes: extracting key information from the demand content and denoising it to obtain a first demand; importing the first demand into the comprehensive database to find similar demands, and adding the corresponding solutions for the similar demands; if no similar demand is found, the first demand is transferred to the artificial customer service for response, and the solutions provided by the artificial customer service are recorded to form tags.

[0017] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: By extracting key information from the demand content and denoising it, the first demand is then used to find similar demands in the comprehensive database. Through the similar demands, the corresponding solutions can be quickly found and tagged, which is convenient for subsequent customers to solve the same problem in a timely manner. At the same time, when no similar demand is found, it is transferred to the artificial customer service, and the solutions are also generated into tags for subsequent quick response to ensure the service for customers.

[0018] In an example of the present invention, in combination with the second portrait, real-time reconstruction and quarterly reconstruction of the first portrait to obtain a third portrait further includes: real-time monitoring of the user's electricity consumption situation to determine whether the user has abnormal electricity consumption behavior; if so, remind the user of the current abnormal electricity consumption, analyze the cause of the abnormal electricity consumption, and find the corresponding solution measures in the comprehensive database to provide to the user and upload and add corresponding tags; if not, continue to monitor; summarize and count the demands put forward by the user quarterly, compare with the quantity threshold, add tags to the solution measures corresponding to the demands exceeding the quantity threshold, and prompt the maintenance personnel that this demand needs to be key checked.

[0019] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: By setting up real-time monitoring of the user's electricity consumption situation, it is possible to send a reminder to the user when there is an abnormal electricity consumption situation, and add labels to solve the abnormality. This can avoid the impact of the user's abnormal electricity consumption on the user's electricity consumption, and can also send a more timely reminder to the user when it occurs later. Through real-time updates of electricity consumption abnormalities, it can ensure that the user portrait more completely conforms to the current user's electricity consumption situation. At the same time, it is also set to summarize the user's needs quarterly and add labels to the solutions to the demands that exceed the quantity threshold, so that the problems most frequently encountered by the user in a quarter can be solved, and the user's electricity consumption needs can be guaranteed.

[0020] The present invention also provides a big data-based power customer portrait management system. The power customer portrait management system is used to implement the power customer portrait management method as described in any one of the above. The power customer portrait management system includes: a collection module for collecting data; a portrait module for portrait of users; and a reply module for providing corresponding solutions and demand replies for users.

[0021] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: By setting up a collection module, it is more convenient to collect data, which is convenient for subsequent portrait generation. At the same time, the portrait module can portrait users more accurately and conveniently. Combining with the reply module to reply to the user's demands can better guarantee the user's electricity consumption.

[0022] After adopting the technical solution of the present invention, the following technical effects can be achieved: (1) By setting up to collect the power data, environmental data, social data, and interaction data of each user, the accuracy and comprehensiveness of the subsequent portrait of the user can be improved through multiple dimensions. The multiple-dimension method can also facilitate subsequent responses to different situations of customers, so as to better provide services for customers. At the same time, it is also set to perform secondary portrait when the current proposed portrait does not meet the customer's demands, and further improve the customer's portrait through the secondary portrait method, so as to better provide services for the customer when encountering the customer next time. At the same time, it also dynamically updates the portrait through real-time reconstruction and quarterly reconstruction, ensuring the real-time effectiveness and accuracy of the portrait, so as to better provide subsequent services for customers and better and more comprehensively manage the customer portrait, guaranteeing better solution of the content demanded by the customer, improving the accuracy and practicality of the portrait, and also making the management method more intelligent. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the accompanying drawings to be used in the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings; Figure 1 It is a flowchart of a method for managing power customer portraits based on big data provided by an embodiment of the present invention; Figure 2 It is a module diagram of a system for managing power customer portraits based on big data provided by an embodiment of the present invention.

[0024] Explanation of reference numerals: 100 is a power customer portrait management system; 110 is a collection module; 120 is a portrait module; 130 is a reply module. Specific embodiments

[0025] To make the above objects, features, and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0026]

First Embodiment

[0027] Specifically, in daily use, the power IoT terminal is used to collect power operation data and environmental monitoring data to obtain power data and environmental data. Among them, the environmental monitoring data can be obtained by the power IoT terminal from the meteorological system. At the same time, the social attribute database and the user behavior log server are accessed to obtain social data and interaction data. Among them, the power data can be meter data, fault records, etc., the environmental data can be meteorological information, new energy power generation prediction, etc., the social data can be geographical information, electricity bill information, etc., and the interaction data can be customer service records, app interaction logs, etc.

[0028] Furthermore, the power data, environmental data, social data, and interaction data of each user are sorted and integrated, and a first virtual image is created for each user to form a virtual portrait. Specifically, through the fault records in the power data, the fault forms and fault contents recorded in the power IoT terminal for each user can be obtained. After denoising and extraction, different fault records are classified and sorted by the number of occurrences and priority, and the fault records with the most occurrences recorded by the user are ranked at the top to obtain the first sequence.

[0029] Furthermore, the first, second, and third virtual image conditions are respectively generated in combination with meteorological information, electricity bill information, and customer service records to generate the virtual portrait of each user. Specifically, the area where the user is located and the corresponding weather conditions are obtained from the meteorological information. At the same time, it is checked from the first sequence whether the user has a power outage fault. If there is a power outage fault and it ranks among the top two, it means that the user has encountered more power outage faults. At the same time, in combination with the meteorological information, it is judged whether there is bad weather such as typhoon, blizzard, etc. that can cause power outages on the date when the user encounters the power outage fault, so as to identify whether the user is in an area where bad weather often occurs, and thus a label of power outage caused by bad weather can be identified for the user, and then a corresponding preventive measure label can be generated. When bad weather is expected in the future, a timely reminder can be sent to the customer, so that the user can take preventive measures in advance. At the same time, if it is not bad weather, it means that there may be a problem with the line in the user area and needs to be checked, and then a measure label for checking the line is added. At the same time, the labels of multiple users can be integrated later. If there are multiple users in the area and they all have the label of checking the line, it means that there is a serious problem with the line in this area and needs to be checked in time, which can better guarantee the electricity use of the local residents.

[0030] Further, check whether there is a power outage fault of individual businesses in the first sequence. If it exists and ranks among the top two, obtain the electricity bill information of the user, check whether the electricity bill balance of the user is insufficient, and determine whether the power outage fault is caused by the user's failure to pay the electricity bill in time. If so, add a label for reminder measures. In the current environment, there are many cases where users forget to submit their electricity bills, resulting in sudden power outages and affecting their lives. Therefore, by identifying the electricity bill balance and generating it as a label, users can be reminded to pay the electricity bill in time. If not, it is also necessary to add a label for regional line inspection measures for this user. At the same time, combined with the geographical location of this user, determine whether there are users with the same label in the same village, the same community, or the same residential building. If so, it is necessary to conduct line inspections in a timely manner to avoid affecting normal power use.

[0031] Further, compare the records in the first sequence by checking the user requirements recorded in the customer service records, so as to determine whether there are cases where the problems that users often seek customer service help rank among the top two. If so, it means that this requirement is often mentioned, indicating that it is still occurring and this appeal has not been resolved. Then, it is necessary to generate corresponding solutions to form labels, and subsequently follow up to solve the problems to ensure the user's power consumption needs.

[0032] Further, combine the labels in the three simulacrum conditions to form the final user simulacrum, so that the user simulacrum is more comprehensive.

[0033] Further, when the user raises an appeal, extract the user's simulacrum as the first portrait, and extract the content of the appeal for denoising. Determine whether it can be solved through the labels recorded in the first portrait. If not, it is necessary to generate a corresponding second simulacrum, extract key information from the content of the appeal and denoise it to obtain the user's first requirement, and import it into the comprehensive database to summarize and find similar requirements. Specifically, there are solutions to different problems encountered by many users in the comprehensive database. Problems that a certain user has not encountered may be encountered by other users. Therefore, the corresponding solution can be found more quickly, and corresponding labels can be generated, and then quickly feedback to the customer to ensure that the customer can obtain the solution more quickly when encountering the same problem later. When no similar requirements are found, it means that a new problem has been encountered, and manual intervention is required to solve it, and the final solution measure is uploaded for subsequent follow-up and for providing solutions to other users' problems.

[0034] Furthermore, in the daily maintenance of subsequent customer portraits, real-time reconstruction and quarterly reconstruction are added to dynamically update the customer portraits. Specifically, in daily life, the electricity consumption of users is monitored in real time. When abnormal electricity consumption of a user is detected, the user is reminded and the cause of the abnormality is analyzed. For example, when the electricity consumption speed of a user exceeds the daily average value, the user is reminded that the current electricity consumption is too fast and asked to confirm whether there is an abnormality in the electrical appliances. When the user feedback indicates that the electrical appliances are normal, there may be a wiring problem that needs to be solved manually, or the user may need to use a certain high-power electrical appliance in this situation. At this time, corresponding tags need to be added to remind the user to pay attention to electrical safety tags, etc.

[0035] Furthermore, by sorting, summarizing, and counting the demands put forward by users on a quarterly basis, and then comparing them with the quantity threshold, when the quantity of a certain demand exceeds the quantity threshold, it indicates that the current problem is encountered by many users and is still occurring, indicating that the demand has not been substantially solved. Then a tag for on-site maintenance is generated, which is convenient for maintenance personnel to quickly identify the tag during daily work, and then conduct priority investigation to ensure the electricity consumption needs of users. Among them, the quantity threshold is a manually set value and can be changed according to the actual situation.

[0036] Preferably, by setting to collect the power data, environmental data, social data, and interaction data of each user, and then improving the accuracy and comprehensiveness of the subsequent user portraits through multiple dimensions. The multi-dimensional method can also facilitate dealing with different situations of customers in the future, and thus can better provide services for customers. At the same time, when the current proposed portrait does not meet the customer's demands, a secondary portrait is set, and the customer's portrait is further improved through the secondary portrait method, so that better services can be provided for the customer when encountering the customer next time. At the same time, real-time reconstruction and quarterly reconstruction are also carried out to dynamically update the portrait, ensuring the real-time effectiveness and accuracy of the portrait, and thus being able to better provide subsequent services for customers and manage the customer portrait more comprehensively. It can ensure that the content demanded by the customer is better solved, improve the accuracy and practicality of the portrait, and also make the management method more intelligent.

[0037] Specifically, a first quasi-portrait is made for each user according to the power data, environmental data, social data, and interaction data, and the quasi-portrait corresponding to each user is obtained, including: the power data includes fault records, the environmental data includes meteorological information, the social data includes electricity bill information, and the interaction data includes customer service records; the quasi-portrait of each user is made by combining the fault records, meteorological information, electricity bill information, and customer service records corresponding to each user.

[0038] Preferably, through the fault records corresponding to each user, it is possible to clearly know the content of the user's previous fault declarations, and then obtain the problems that the user often faces and needs to solve, so as to facilitate subsequent profiling. Through meteorological information, it is possible to clearly know whether the user's location encounters bad weather that is likely to cause power failures, so as to better help judge the problems encountered by the user when reporting faults, ensure quick identification of problems and serve the user. At the same time, setting electricity bill information can quickly judge the user's electricity bill situation, so as to avoid power problems caused by electricity bill issues. Furthermore, by setting customer service records as a factor, the customer service records can reflect most of the user's demands, and thus the combined profile can be more accurate.

[0039] Specifically, the process of creating a preliminary profile for each user by combining the fault records, meteorological information, electricity bill information, and customer service records corresponding to each user also includes: classifying the faults in the fault records and sorting them by the number of records to obtain a first sequence; combining the meteorological information and the first sequence to extract the first preliminary profiling condition; combining the electricity bill information and the first sequence to extract the second preliminary profiling condition; combining the customer service records and the first sequence to extract the third preliminary profiling condition; and creating a preliminary profile for each user according to the first, second, and third preliminary profiling conditions.

[0040] Preferably, by setting the classification of the fault records and then performing priority sorting, it is possible to quickly integrate the user's faults, identify the problems that the user often encounters, and then quickly perform label recognition and establishment, so as to better serve the user. At the same time, by combining meteorological information, electricity bill information, and customer service records respectively to extract different preliminary profiling conditions, the finally obtained preliminary profile can be more accurate and comprehensive, ensuring the integrity of the profile.

[0041] Specifically, the process of combining the meteorological information and the first sequence to extract the first preliminary profiling condition also includes: determining whether there is a power outage fault in the first sequence; if it exists and ranks among the top two in the first sequence, then combining the meteorological information to judge whether the power outage is caused by bad weather; if the judgment is yes, then add a label indicating that bad weather affects the power outage and that preventive measures need to be informed to the corresponding user in advance when bad weather is encountered; if the judgment is no, then add a label indicating that the line in the corresponding user's area needs to be checked.

[0042] Preferably, by extracting the power outage fault and combining the meteorological information to judge the specific cause of the power outage fault, it is possible to quickly add preventive measure labels to the user, and then inform the user in advance when there is bad weather in the future in the user's subsequent location, improving the user's risk resistance ability and better serving the user. At the same time, when it is not affected by bad weather, add a label for checking the regional lines for the user, so as to further avoid the user's power outage risk and ensure the normal power supply of the user.

[0043] Specifically, combining the electricity fee information and the first sequence, the extraction of the second quasi-image condition further includes: determining whether there is an outage fault of self-employed households in the first sequence; if it exists and ranks among the top two in the first sequence, then combining the electricity fee information to determine whether the outage is caused by insufficient electricity fee balance; if so, add a measure label for reminding the corresponding user to pay the electricity fee; if not, add a measure label for inspecting the lines in the area where the corresponding user is located.

[0044] Preferably, by setting to judge the outage fault of self-employed households in combination with the electricity fee balance, it is possible to prevent the user from having an outage fault due to failure to submit the electricity fee in time, and then remind the user early, so that the user can pay the electricity fee in time, thereby ensuring that the user can use electricity normally.

[0045] Specifically, combining the customer service record and the first sequence, the extraction of the third quasi-image condition further includes: determining whether the user requirements recorded in the customer service record exist in the first sequence; if it exists and ranks among the top two in the first sequence, then generate a label for the solution corresponding to the requirement record, and provide subsequent follow-up service to the user corresponding to the requirement record.

[0046] Preferably, by extracting the content in the customer service record and comparing it with the first sequence, it is possible to add records of the demands that the user asks the customer service more frequently, and generate corresponding labels for the solutions provided by the customer service at that time, so as to facilitate direct intelligent reply when encountering the same problem subsequently, and be able to provide subsequent follow-up service to ensure the user's electricity demand.

[0047] Specifically, if not, perform a second quasi-image for the current user to obtain a second portrait. Solving the appeal content according to the second portrait further includes: extracting key information from the appeal content and denoising it to obtain a first requirement; importing the first requirement into the comprehensive database to find similar requirements, and adding the corresponding solution measures for the similar requirements; if no similar requirement is found, transfer the first requirement to the artificial customer service for response, and record the solution provided by the artificial customer service to form a label.

[0048] Preferably, by extracting key information from the appeal content and denoising it, then finding similar requirements for the first requirement in the comprehensive database, it is possible to quickly find the corresponding solution measures through the similar requirements and add labels, which is convenient for subsequent customers to solve the same problem in time. At the same time, when no similar requirement is found, transfer it to the artificial customer service, and generate labels for the solution at the same time, which is convenient for subsequent quick response and ensures the service to the customer.

[0049] Specifically, in combination with the second image, real-time reconstruction and quarterly reconstruction of the first image are performed to obtain the third image, which further includes: real-time monitoring of the user's electricity consumption to determine whether the user has abnormal electricity consumption behavior; if so, the user is reminded of the current abnormal electricity consumption, the reasons for the abnormal electricity consumption are analyzed, and corresponding solutions are searched in the comprehensive database and provided to the user and uploaded with corresponding tags added; if not, continuous monitoring is carried out; the demands put forward by the user are summarized and counted quarterly, compared with the quantity threshold, the solutions corresponding to the demands exceeding the quantity threshold are tagged, and the maintenance personnel are prompted that this demand needs to be key checked.

[0050] Preferably, by setting to monitor the user's electricity consumption in real time, it is possible to remind the user when there is abnormal electricity consumption, and add tags to solve the abnormality, thereby avoiding the impact of the user's abnormal electricity consumption on the user's electricity consumption, and being able to remind the user more timely when it occurs later. Through the real-time update of the electricity consumption abnormality, it can ensure that the user's portrait more completely conforms to the current user's electricity consumption situation. At the same time, it is also set to summarize the user's needs quarterly, and add tags to the solutions of the demands exceeding the quantity threshold, so as to solve the problems that the user encounters most in a quarter and ensure the user's electricity consumption needs.

[0051] See Figure 2 As shown in, the present invention also provides a big data-based power customer portrait management system 100. The power customer portrait management system 100 is used to implement the power customer portrait management method as described in any one of the above. The power customer portrait management system 100 includes: a collection module 110 for collecting data; a portrait module 120 for creating a portrait of the user; and a reply module 130 for providing corresponding solutions and demand replies for the user.

[0052] Preferably, by setting the collection module 110, it is more convenient to collect data, which is convenient for subsequent portrait generation. At the same time, the portrait module 120 can create a portrait of the user more accurately and conveniently. Combining with the reply module 130 to reply to the user's demands can better ensure the user's electricity consumption.

[0053] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for managing power customer portraits based on big data, characterized in that, Including: Real-time collection of power operation data and environmental monitoring data through the power Internet of Things terminal, and at the same time accessing the social attribute database and the user behavior log server, and then establishing a comprehensive database; Obtaining power data, environmental data, social data and interaction data of each user according to the comprehensive database; Performing a first simulation on each user according to the power data, the environmental data, the social data and the interaction data to obtain a corresponding portrait for each user; Obtaining the content of the current user's appeal, obtaining the corresponding portrait of the current user and recording it as the first portrait; Judging whether the content of the appeal falls into the first portrait; If not, perform a second simulation on the current user to obtain a second portrait, and solve the content of the appeal according to the second portrait; Combining the second portrait, performing real-time reconstruction and quarterly reconstruction on the first portrait to obtain a third portrait, and using the third portrait to replace the first portrait, and following up the subsequent appeals of the current user through the third portrait; If so, reply and solve the content of the appeal according to the measures recorded in the first portrait.

2. The method for managing power customer portraits based on big data according to claim 1, wherein The performing a first simulation on each user according to the power data, the environmental data, the social data and the interaction data to obtain a corresponding portrait for each user includes: The power data includes fault records, the environmental data includes meteorological information, the social data includes electricity bill information, and the interaction data includes customer service records; Combining the corresponding fault records, meteorological information, electricity bill information and customer service records of each user to perform the portrait for each user.

3. The method for managing power customer portraits based on big data according to claim 2, wherein, The combining the corresponding fault records, meteorological information, electricity bill information and customer service records of each user to perform the portrait for each user further includes: Classifying the faults in the fault records and sorting them by the number of records to obtain a first sequence; Combining the meteorological information and the first sequence to extract the first simulation condition; Combining the electricity bill information and the first sequence to extract the second simulation condition; Combining the customer service records and the first sequence to extract the third simulation condition; Performing the portrait for each user according to the first, second and third simulation conditions.

4. The method for managing power customer portraits based on big data according to claim 3, wherein The combining the meteorological information and the first sequence to extract the first simulation condition further includes: Judging whether there is a power outage fault in the first sequence; If it exists and ranks among the top two in the first sequence, combine the meteorological information to judge whether the power outage is caused by bad weather; If the judgment is yes, add a label indicating that the power outage is affected by bad weather and that preventive measures need to be informed to the corresponding user in advance when encountering such bad weather; If the judgment is no, add a label indicating that the power lines in the corresponding user's area need to be checked.

5. The method for managing power customer portraits based on big data according to claim 4, wherein The combining the electricity bill information and the first sequence to extract the second simulation condition further includes: Judging whether there is an individual household power outage fault in the first sequence; If it exists and ranks among the top two in the first sequence, combine the electricity bill information to judge whether the power outage is caused by insufficient electricity bill balance; If so, add a label indicating that the corresponding user needs to be reminded to pay the electricity bill; If not, add a measure label that requires line troubleshooting for the region where the corresponding user is located.

6. The method for managing power customer portraits based on big data according to claim 5, characterized in that The extraction of the third simulacrum condition in combination with the customer service record and the first sequence further includes: Determine whether the user requirements recorded in the customer service record exist in the first sequence; If it exists and ranks among the top two in the first sequence, generate a label for the solution corresponding to the requirement record, and provide subsequent follow-up services to the user corresponding to the requirement record.

7. The method for managing power customer portraits based on big data according to claim 1, wherein The "if not" mentioned above, then perform a second simulation on the current user to obtain a second portrait. The solution to the appeal content according to the second portrait further includes: Extract key information from the appeal content and denoise it to obtain the first requirement; Import the first requirement into the comprehensive database to find similar requirements, and add the corresponding solution measures of the similar requirements; If no similar requirements are found, transfer the first requirement to the artificial customer service for response, and record the solution provided by the artificial customer service to form a label.

8. The method for managing power customer portraits based on big data according to claim 1, wherein The real-time reconstruction and quarterly reconstruction of the first portrait in combination with the second portrait to obtain a third portrait further includes: Real-time monitor the user's electricity consumption situation and determine whether the user has abnormal electricity consumption behavior; If it exists, remind the user of the current abnormal electricity consumption, analyze the reason for the abnormal electricity consumption, and find the corresponding solution measures in the comprehensive database to provide to the user and upload and add the corresponding labels; if not, continue to monitor; Summarize and count the appeals made by the user on a quarterly basis, compare with the quantity threshold, add labels to the solution measures corresponding to the appeals exceeding the quantity threshold, and prompt the maintenance personnel that this appeal needs to be key checked.

9. A power customer portrait management system based on big data, the power customer portrait management system is used to implement the power customer portrait management method according to any one of claims 1-8, characterized in that, The power customer portrait management system includes: A collection module, which is used to collect data; A portrait module, which is used to portrait users; A reply module, which is used to provide corresponding solution measures and demand replies to users.

Images